← Back to community benchmarks

Qwen3-Coder-30B-A3B-Instruct-4bit-dwq-v2

M2 Pro (16c) · 32 GB · 4bit · 2026-05-18

Performance

1k

tokens

262.6

PP tok/s

50.4

TG tok/s

3899

TTFT (ms)

16.6

Peak mem (GB)

Hardware

Chip M2 Pro (16c)

Memory 32 GB

GPU Cores 16

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	262.6	50.4	16.6 GB	current
4k	276.2	37.1	17.2 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	50.4	1.00×
2×	56.2	1.12×
4×	61.9	1.23×